| { | |
| "best_metric": 0.6076175570487976, | |
| "best_model_checkpoint": "bert_tiny_lda_20_v1_book_cola/checkpoint-102", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 272, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8408417105674744, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6134, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6152606010437012, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3459, | |
| "eval_samples_per_second": 3015.731, | |
| "eval_steps_per_second": 14.457, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.9225541353225708, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6043, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6082664132118225, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3484, | |
| "eval_samples_per_second": 2993.443, | |
| "eval_steps_per_second": 14.35, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.8375650644302368, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5889, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6922339200973511, | |
| "eval_loss": 0.6076175570487976, | |
| "eval_matthews_correlation": 0.0463559874942472, | |
| "eval_runtime": 0.357, | |
| "eval_samples_per_second": 2921.64, | |
| "eval_steps_per_second": 14.006, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.4006083011627197, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5597, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6922339200973511, | |
| "eval_loss": 0.6156318187713623, | |
| "eval_matthews_correlation": 0.11271502960566541, | |
| "eval_runtime": 0.3603, | |
| "eval_samples_per_second": 2894.472, | |
| "eval_steps_per_second": 13.876, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.4112930297851562, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.509, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6720997095108032, | |
| "eval_loss": 0.6309717297554016, | |
| "eval_matthews_correlation": 0.18791963250554952, | |
| "eval_runtime": 0.363, | |
| "eval_samples_per_second": 2873.273, | |
| "eval_steps_per_second": 13.774, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.682448148727417, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.4586, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6816874146461487, | |
| "eval_loss": 0.6527789235115051, | |
| "eval_matthews_correlation": 0.1793431256335088, | |
| "eval_runtime": 0.3494, | |
| "eval_samples_per_second": 2984.813, | |
| "eval_steps_per_second": 14.309, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.6643636226654053, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.415, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6577181220054626, | |
| "eval_loss": 0.6662452816963196, | |
| "eval_matthews_correlation": 0.12733350576613764, | |
| "eval_runtime": 0.3411, | |
| "eval_samples_per_second": 3057.43, | |
| "eval_steps_per_second": 14.657, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.269206523895264, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.3704, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.666347086429596, | |
| "eval_loss": 0.7398048043251038, | |
| "eval_matthews_correlation": 0.13162759224670714, | |
| "eval_runtime": 0.3314, | |
| "eval_samples_per_second": 3146.956, | |
| "eval_steps_per_second": 15.086, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 1793902531485696.0, | |
| "train_loss": 0.5149053896174711, | |
| "train_runtime": 49.4041, | |
| "train_samples_per_second": 8654.135, | |
| "train_steps_per_second": 34.41 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 1700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1793902531485696.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |